Three‐dimensional velocity structure of the northern Hikurangi margin, Raukumara, New Zealand: Implications for the growth of continental crust by subduction erosion and tectonic underplating

Abstract

Traveltimes between shots from nine marine seismic reflection lines and nine onshore recorders were used to construct a 3‐D P wave velocity model of the northern Hikurangi subduction margin, New Zealand. From north to south between Raukumara Basin and Raukumara Peninsula, the Moho of the overriding plate increases in depth from 17 to ∼35 km. Low seismic P wave velocities of 3.5–5.0 km/s are localized within a ∼10 km thick prism in the lower crust of the overriding plate immediately updip of the intersection between the subduction thrust and Moho and beneath the topographic crest of East Cape Ridge and the Raukumara Range. Southward, this region of low seismic velocities and surface uplift increases in distance from the trench as the thickness of the crust in the overriding plate increases. We interpret this low‐velocity volume to be underplated sedimentary rocks and crustal materials that were tectonically eroded by subduction beneath the trench slope. The buoyancy and low strength of these subducted materials are proposed to assist the escape from a subduction channel near the base of the crust and drive local rock uplift. In the middle crust, our observations of very low velocity suggest high fluid‐filled porosities of 12%–18%, and the implied buoyancy anomaly may enhance underplating. At greater depths the process is driven by the contrast between upper crustal quartz‐feldspar mineralogy and the denser diabase or olivine‐rich lithologies of the lower crust and mantle. We estimate a rate of lower crustal underplating at the northern Hikurangi margin of 20 ± 7 km3 Ma−1 km−1 since 22 Ma. We suggest that underplating provides an efficient means of accreting subducted sediment and tectonically eroded material to the lower crust and that the flux of fore‐arc crustal rocks into the mantle at subduction zones may be systematically overestimated.